Amelogenesis imperfecta: Human amelogenin gene mutations result in amelogenesis imperfecta (AI), the most common hereditary condition affecting enamel, which is characterized by thin and/or poorly mineralized enamel. Our earlier studies demonstrated that amelogenin null (Amel-/-) mice have an enamel phenotype that appears similar to that observed in human AI, confirming the important role of amelogenins in normal development of enamel. Subsequently, we demonstrated that 2 amelogenin isoforms, a full-length M180 amelogenin and a leucine-rich amelogenin peptide (LRAP), are expressed in cementum, and their absence in aging Amel-/- mice is associated with cementum defects. We further characterized the functions of the amelogenin isoforms in osteoclastogenesis and in the proliferation and migration of cementoblast/periodontal ligament cells. Our studies provided strong evidence that LRAP inhibits osteoclastogenesis We further analyzed amelogenin functions in bone metabolism by generating transgenic mice that express amelogenin spliced variants in bones. Our initial analysis revealed that amelogenins may play a role in maintenance of bone metabolism. In collaboration with the Molecular Biology Section, LCDB, NIDCR, we developed ameloblastin knockout mice that have severe enamel hypoplasia, detached ameloblasts, and odontogenic tumors of epithelial origin. In order to search for synergistic roles of amelogenin and ameloblastin in enamel development, we generated amelogenin and ameloblastin double-null (Amel-/-/Ambn-/-) mice. These mice showed more severe enamel defects than Amel-/- mice or Ambn-/- mice. SEM analysis showed that enamel structure was completely lost, and the ameloblast layer was irregular and detached from the basement membrane in Amel-/-/Ambn-/- mice. Proteomic analysis revealed the presence of an increased level of Rho-GDI only in Amel-/-/Ambn-/- tooth buds, suggesting synergistic effects of amelogenin and ameloblastin. [unreadable] Dentinogenesis Imperfecta: DSPP forms a major constituent of dentin extracellular matrix proteins and is believed to play important roles in the mineralization process that forms mature dentin. Several mutations have been identified in the DSPP gene in patients with dentinogenesis imperfecta. DSPP is predominantly expressed in dentin-producing odontoblasts and transiently in enamel-producing ameloblasts. Low levels have also been detected in several other tissues like bone, inner ear, salivary glands, and kidneys. To gain insights into the molecular roles of DSPP in dentinogenesis, we previously generated DSPP-/- mice. The structural tooth defects observed in these mice were enlarged pulp chambers, increased width of predentin zone, hypomineralization, pulp exposure, irregular mineralization front, and a lack of uniform coalescence of calcospherites in the dentin. The levels of the proteoglycans biglycan and decorin were increased in the widened predentin zone and in the void spaces among the calcospherites in the dentin of DSPP null mice. These enhanced levels correlated well with the regions defective in mineralization and further indicated that these molecules may adversely affect the dentin mineralization process by interfering with the coalescence of calcospherites. However, type I collagen levels were unaffected in the null teeth. Therefore, we hypothesized that the increased levels of biglycan and decorin in the DSPP-/- mice interact with collagen fibrils and promote maturation, but they fail to dissociate from the mature collagen, which is required for subsequent dentin mineralization. We proposed that DSPP or its cleaved peptides, in addition to their suggested role in nucleation of mineralization, may play a pivotal role in the regulation of biglycan and decorin levels during dentinogenesis and together may form the basis for the dentin defects seen in the DSPP-/- mice. In order to understand the molecular mechanism underlying this phenotype, we first examined whether the elevated levels of biglycan and decorin are causative factors or elevated as a consequence. Towards this goal, we have generated 2 mouse models: DSPP-/-;biglycan-/- and DSPP-/-;decorin-/-. Detailed analysis of these 2 mouse models indicates that the deficiency of decorin but not biglycan rescues the enlarged predentin phenotype of DSPP-/- mice. Increased levels of decorin in DSPP-/- predentin contribute to abnormal enlargement of predentin. The DSPP mRNA is translated into a single protein, DSPP, which is cleaved into 3 peptides, dentin sialoprotein (DSP), dentin glycoprotein (DGP), and dentin phosphoprotein (DPP). We constructed transgenic vectors that express DSP/DGP under the control of the DSPP promoter and generated independent transgenic mouse lines that express DSP/DGP in odontoblasts and preameloblasts. Two of these lines were bred with DSPP-/- mice to establish 2 mouse lines that express DSP/DGP at low and high levels in the DSPP null background (TgDSPKO). An initial analysis revealed that the spatial and temporal expression patterns of the DSP/DGP transgene are similar to that of endogenous DSPP, and there is a partial rescue of the null dentin phenotype, suggesting that DSP/DGP and DPP have specific roles in dentin mineralization.[unreadable] Role of TGF-beta signaling pathway in tooth development and disease: Of these 3 isoforms, only TGF-beta1 is expressed throughout tooth development, but its specific role in tooth biology is far from clear. We have previously reported our first strategy based on DSPP-TGF-beta1 transgenic mice and showed that these mice had dentin defects similar to those observed in DGI-II and dentin dysplasia. Our further investigations revealed that the transient TGF-beta1 expression in the presecretory ameloblasts results in detachment of secretory ameloblasts from the dentin surface, and the detached ameloblasts form cyst-like structures adjoining dentin. The enamel appeared aprismatic and amorphous along with the presence of mineralized clusters of globules in the cysts. In a second approach to disrupting TGF-beta signaling in the odontoblasts, we generated odontoblast-specific conditional knockout mice for TGF-beta Receptor I and TGF-beta Receptor II. These conditional knockout mice were recently generated using a well-characterized DSPP-Cre mouse line that showed significant deletion of receptor expression in the odontoblasts but did not display any apparent phenotype. Because TGF-beta signaling has been implicated in pulpitis and secondary dentinogenesis, we followed a unique strategy to test its role. In collaboration with the Forsyth Institute, Boston, we drilled a small cavity through the enamel touching the dentin-enamel junction (DEJ) of the first mandibular molars of these mice and their controls and filled this cavity with a bonding material that contained LPS (to mimic bacterial infection) or PBS (control). Our preliminary analysis revealed focal inflammation and increased degeneration of the tooth pulp in the conditional knockout molars implanted with LPS, suggesting an important role of TGF-beta signaling in the immune response, and in resolution of inflammation.[unreadable] [unreadable] Distribution of efforts: Amelogenesis imperfecta: 30%; Dentinogenesis imperfecta: 70%